math.NAcs.LG
Physics-informed traditional and generative Mori-Zwanzig-reduced models for multiscale systems
📅
👥 4 authors
arXiv:2611.05211v1
Authors
Emily WilliamsSaad QadeerAmanda HowardPanos Stinis
Abstract
Model reduction via the Mori-Zwanzig (MZ) formalism is well understood for coarse-grained representations constructed from standard basis functions such as Fourier functions. We show that the same formalism can be used to build reduced-order models in bases discovered by physics-informed machine learning. The fluctuation-dissipation theorem (FDT) is used to close the resulting non-Markovian evolution equations for the expansion coefficients of the learned basis. We further introduce a complementary, generative closure model. Relative to the Markovian dynamics alone, both the MZ and generative closures substantially improve predictions of the filtered states.
Paper Information
- arXiv ID:
- 2611.05211v1
- Published:
Related Categories
math.NAcs.LG